A+ A A-

Consensus decision-making

Consensus decision making

Consensus decision making refers to a democratic decision process that not only seeks the agreement of a majority of participants, but also to resolve or mitigate the objections of the minority to achieve the most agreeable decision.

The method tends to deemphasize the role of factions or parties and promote the expression of individual voices.

The method also increases the likelihood of unforeseen or creative solutions by juxtaposing dissimilar ideas.

Because it seeks to minimize objection, it is most popular with voluntary organizations, wherein decisions are more likely to be carried out when they are most widely approved.

Consensus methods are desirable when enforcement of the decision is unfeasible, such that every participant will be required to act on the decision independently.

Minority views must be considered to a greater degree than in circumstances where a majority can take the action and enforce the decision without any further consultation with the minority voters.

Since higher levels of consensus usually require more time and effort to achieve, groups may reserve consensus decision methods for particularly complex, risky or important decisions.

Rather than simply list known alternatives, debate for a short time, vote, and then accept or reject by some percentage of majority (say 50% plus one, or 2/3), a consensus decision making process involves generating new alternatives, combining elements of multiple alternatives, and in general spending considerably more time in debate and in checking that people understand a proposal or an argument.

This empowers minorities, those with objections that are hard to state quickly, and those who are ineloquent in debate.

Therefore consensus decision making is considered a form of grassroots democracy.

Consensus decisions are considered desirable by all egalitarian groups that seek to reduce the amount of power delegated to leaders, chairpersons or agenda setters and the amount of harm or loss than can be imposed on minorities (or individuals) by a majority.

Consensus methods are appropriate when personal (or emotional) risk to members is high, trust is low, and time is available for a prolonged discussion. Consensus methods may be used to remedy patterns of decision making based on habit, subservience or carelessness.

Like any group decision making, consensus decision making can disempower those not present in the debating forum, as they cannot expect to have input on the new measures that are proposed (whereas they could have had considerable time for input into the known alternatives prior to the debate).

Accordingly, most systems of consensus decision making place a premium on participation, empowering those whose alternative time uses are less attractive or remunerative.

Three key issues tend to define a particular type of consensus decision making: degree of agreement or unanimity required, timing of presentation including division of time among urgent versus important matters, and

followup to action including the monitoring that arises from dissent, and from claims of majority proponents whose preferred course of action is being taken over minority objections.

There is also the question of facilitation or process leadership, which is handled separately at the end of this article.

Consensus is not unanimous: so who must agree?

A healthy consensus decision making process usually encourages and outs dissent early, maximizing the chance of accommodating the views of all minorities.

It also usually assigns a role to the dissenter, e.g. the Vatican assigns the role of devil's advocate to a specific priest who argues against beatification of a saint, to ensure the case 'against' remains well represented. After the decision, the dissenting minority may have some role to play in monitoring the decision.

In the Supreme Court of the United States, both the majority opinion and minority opinion are equally well documented, as the legal grounds for agreeing with either may exist in some court in future.

Differing degrees of consensus provide differing group dynamics.

If unanimity is rejected, then various definitions of consensus can be invoked, most easily defined as follows: Unanimity: Many groups consider unanimous decisions a sign of agreement, solidarity, and unity.

However, there is considerable and difficult-to-controvert evidence that most unanimous decisions are a sign of coercion, fear, undue persuasive power or eloquence, inability to comprehend alternatives, or plain impatience with the process of debate.

The ancient Jewish Sanhedrin courts rejected all unanimous verdicts and let the accused go free, on the assumption that the accused had not had any adequate defense.

The New Testament records the hastily-convened illegal Sanhedrin of Pharisees that condemned Jesus as a unanimous verdict - one unacceptable under Torah law. Such practices were actually criticized by Jesus and other contemporary rabbis, and many believe they reflected the influence of more authority-driven Roman law (on which our current jurisprudence in the Western world is based).

"Unanimity minus one" or U-1, requires all delegates but one to support the decision.

The individual dissenter cannot block the decision although they may be able to prolong debate (e.g. via a filibuster).

The dissenter may be the ongoing monitor of the implications of the decision, and their opinion of the outcome of the decision may be solicited at some future time.

Betting markets in particular rely on the input of such lone dissenters, which discipline the market's odds, and profit handsomely when they correctly foresee the future.

A betting market can reasonably be considered a U-1 consensus system on the odds themselves: the lone bettor against the odds profits when his or her prediction of the outcomes proves to be better than that of the majority.

Even the simple majority-rule systems tend to invoke U-1 rules when setting agendas, that is, a single individual cannot propose a measure, it must be 'seconded' to be heard out.

"Unanimity minus two" or U-2, does not permit two individual delegates to block a decision, nor one, but it curtails debate with a lone dissenter more quickly so that dissenting pairs can present multiple (hopefully different) views of what is wrong with the decision the majority is contemplating.

By focusing on a pair of dissenters, and allocating less time to lone wolves or 'consensus thugs', a U-2 system tends to form stronger bonds among those who find themselves 'alone on an island with' each other.

Pairs of delegates should be empowered to cut deals, make ethical tradeoffs, and otherwise find the common ground that will enable them to convince a third, decision-blocking, voter to join them.

If two individual delegates cannot convince a third delegate in some finite period of time to join them, then their arguments are considered to be unconvincing or immature or self-interested.

However, the consensus process of U-2 has maximized their chance to get to know each other and to cooperate, so a truly effective presentation and marshalling of support may be possible later on.

Venture capitalists tend to value these traits, and so most will consider hearing a business plan only if it is presented by a pair of business partners already committed to working with each other.

Since only a tiny minority (under 1%) of business plans heard are funded, the experience of joint development of an entrepreneurial opportunity (which itself must challenge majority views of value prevalent in the market, or it fails) is considered to be very important.

Venture capitalists often remark after losing money that they do not regret it, as they learned and gained a great deal from having funded a particular pair of business partners.

If two people dissent against some common measure, it's more likely that the discussion between them can be extended to third parties easily, since it is already verbalized and illustrated. Western European and derived court systems recognize this by strongly encouraging criminal defendants or civil plaintiffs to get an attorney's aid, so that their case can be fully heard out long before the decision.

"Unanimity minus three", or U-3, and other such systems tend to simply recognize the ability of three or more delegates to actively block or sabotage a decision; A decision is moot if sufficient opposition is present to block implementation; However, there is controversy over whether a small group of dissenters, e.g. militants or terrorists, is fundamentally morally different than a large minority, e.g. an opposition party with support of double-digit percentages of the population.

Accordingly U-3 and lesser degrees of consensus are usually lumped in with statistical measures of agreement: "80%, mean plus one sigma, Two-Thirds, 50% plus one" levels of agreement are simply statistical measurements of the number of delegates (or voters that they represent by proxy) agreeing.

They imply a smaller group that disagrees with the decision, or abstains, and which may have various rights to appeal or to overturn a decision later, perhaps even before it is actually implemented.

A majority rule system usually has at least some measures where executive veto or larger agreement (typically 2/3) is required for implementation of any measure.

For instance, the President of the United States can veto a bill presented by (and agreed by a simple majority vote in each house of) Congress, but the Senate can override his veto by a 2/3 majority vote.

Thus there is considerable effort put into anticipating vetos, levels of party discipline or factional dealing... as a result the Senate very often resembles a combination market and consensus decision making forum, with members trading off votes to establish support for measures they truly care about, giving up power on measures they don't care as much for.

As this example suggests, timing can count for as much as voting.


The quality of alternatives considered is, all else being equal, proportional to the amount of time spent gathering and comparing and combining them.

The term deliberative democracy reflects the deliberation that underlies all good consensus decision making. Jon Elder, Ralph Nader, and others have advocated deliberative measures to extend the time for 'sober second thought'.

A fictional example of deliberative democracy is the Entmoot from J. R. R. Tolkien's novel The Two Towers, the second part of The Lord of the Rings.

The Ents, who are large ancient living intelligent trees, spend days discussing the issue of whether to go to war, in their verbose and many-syllabled language.

It provided an example of a decision wherein individual risk was high, and coercive force was difficult or lacking, therefore suited to consensus methods.

Timely decisions are important.

In many cases, a wrong decision taken in time can be better than a good decision taken later.

Agenda forming and presentation of issues at the right time, when there is sufficient time for their debate, and less urgent issues can take a back seat in debate, is a key responsibility of a leader of any decision making process, but particularly true in consensus styles.

Measures to put items on the agenda, or deny them time on the agenda, and especially deadlines for changes to the agenda (e.g. can the agenda itself be changed during the meeting?), are particularly critical in consensus styles.

To achieve a balance between urgency and importance, it's common to reserve some considerable amount of time for matters that are not urgent, but important at all times, e.g. the decision process itself, which takes care to maintain. Consensus decision processes tend to accelerate, as rising trust over the course of the meeting, combined with fatigue, increase individual tolerance and the cost of dissent.

Placing difficult agenda items first tends to speed a meeting, with the risk that important, but less complex decisions will not be achieved before adjournment.

Decisions about when to split up into 'working groups', how to handle agendas, how to deal with changes to agendas or working groups 'from the floor', etc., are all about the allocation of the group's time to urgent versus important matters.

No consensus decision process can survive without close attention to all these procedural matters, and hearing out issues of safety, fairness, and closure that arise from their application in practice.

Action, monitoring and followup

Action is the point of decision.

Without action, the decision is just talk.

Military leaders from Alexander the Great to the present have emphasized that orders simply do not get carried out unless they are personally followed up by the commander.

The same applies to group decisions, perhaps even more so.

Obviously, the opposing minority is not going to do a good job ensuring that a measure is carried out, although they can ensure that problems that arise from it are well-documented, and that inconveniences of its implementation are contained.

However, they can also take steps to ensure that the inconvenience of implementation is maximized, so as to make the point that the measure was impractical and ill-advised from the beginning.

A good example is the Canadian federal government gun registry, mandated by Bill C-68, and opposed by hunting and other lawful firearms owners.

While protesting the bill and predicting a disaster, these opposition groups encouraged non-compliance with the law, which played some role in ballooning the costs of the registry, making it infeasible.

Such tactics can backfire, as proponents of the measure can argue that without the active opposition, the measure would have been practical and affordable.

A major issue in consensus decisions is whose view of the actual outcome to trust, and who to permit time to present their view.

Consensus decisions are especially vulnerable to sabotage of all kinds, so the assignment of action roles, monitoring (from the original majority and minority opinion to some future time when the results of both sets of predictions can be debated), and other followup (e.g. assessing support of the public for a party after it has taken and publicized a particular measure), is a key responsibility of consensus decision leaders.


Aside from these abstract factors, one must consider the practical matter of the facilitation process.

A hierarchical point of view is that leadership or management of the process (as opposed to leadership of a faction or party) is required.

The role of a facilitator in a consensus decision making process can be much more difficult than that of a simple-majority-party leader if group members distrust each other or unconsciously use manipulative techniques.

For a proponent of any given alternative, reducing objections to their plan by eliciting information or preferences from proponents of other alternatives is difficult if people distrust each other; maniuplative opponents will find it advantageous to misrepresent their concerns or refuse to negotiate.

For these reasons, consensus processes usually require trust among participants and skilled, patient facilitators able to synthesise the state of a proposal.

An argument against consensus decision is that few motivated facilitators are willing to assign themselves a role guiding processes rather than pursuing and promoting specific measures empowering themselves.

As Dee Hock described his role at Visa International, an organisation focussed on profit-making, it was something that anyone could do, but almost no one learnt to do well, and which was largely thankless.

Similar sentiments have been echoed by many "leaders" of organizations committed to peace, ecology, and social justice, which tend to have diffuse benefits, and concentrated costs (the opposite of the factors leading to a tragedy of the commons, a key issue in political economy).

However, leaderless organisations committed to peace, ecology, and social justice, where trust builds up and where different participants are encouraged to learn facilitation skills, find that consensus decision making is a practical and powerful tool.

Some organizations have abandoned consensus decision making for simple majority, judging that the difficulty of building a process to formally weigh all of these factors is not worth it, and that these factors can be handled better informally (i.e. in offline discussions before and after debate) than through the process of consensus itself, at the risque of creating a de facto clique that makes the real decisions.

Before considering any consensus decision making process, a group would be wise to consider the feasibility of building up sufficient trust among participants and the willingness of participants to learn facilitation skills, and whether or not these are compatible with the operational structure of the organisation.

For example, an organisation with a President who hierarchically controls operations could only be compatible with consensus decision making if the President sincerely respected the consensus decision making process.

It would also be intrinsically difficult for a competitive organisation to use consensus decision making, since consensus is a cooperative process, not a competitive process.